Patrick McLeod
> I have a question regarding reporting the estimation results
> of multiple models in a single table. Say I have four
> different models I estimate using -regress-. I store each of
> these models after estimation using -est store modelx- where
> x=1 through 4 for each of my four models.
>
> Is there a way I can integrate each of these model's
> estimation reports (the F, Prob > F, # of obs., R^2, root
> MSE, and independent variable statistics) into a single table
> of output using the four stored models and the -tabulate- or
> -tabstat- command?
>
> Or do I need to just whip up a table in Open Office (or
> Microsoft Word would be comparable) and type them all in by hand?
You never need to type them in by hand. The question is
whether the alternatives are easier, as may not be the
case for a small number of models.
In addition to other favourites in various longitudes, one
possibility is -makematrix- from SSC. That can't pick up
Prob > F, as it's not an e() leftover, but it can do this,
and some more. (Naturally, control over # of dp is available.)
. makematrix , from(e(N) e(F) e(rmse) e(r2)) rhs(headroom-gear) : regress mpg
N F rmse r2
headroom 74 14.876037 5.3033778 .17123291
trunk 74 36.800962 4.7389945 .33824114
weight 74 134.61824 3.4388896 .65153125
length 74 124.327 3.5278698 .63326492
turn 74 77.139026 4.047687 .51722898
displacement 74 71.406715 4.1277922 .49793146
gear_ratio 74 44.068099 4.5882401 .37967451
But the most versatile alternative, ultimately, is to
devise your own output routine with say a -foreach- loop.
Nick
[email protected]
*
* For searches and help try:
* http://www.stata.com/support/faqs/res/findit.html
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/
